Two adjacent candles can overlap through their wicks while their bodies remain separated. That small difference is easy to miss when you are searching for large displacement candles and obvious fair value gaps. In ICT terminology, the separation between those bodies is a volume imbalance. Understanding it begins with reading the open and close correctly, rather than adding a volume indicator to the chart.
The name creates two common mistakes. Some traders assume the pattern measures an imbalance between actual buying and selling volume. Others treat it as a smaller version of a fair value gap and apply identical rules to every touch. Neither shortcut explains what has been observed. A volume imbalance is a specific price feature, and its usefulness depends on the surrounding delivery, the chosen liquidity objective, and the rules used to trade it.
This guide shows how to measure the feature, distinguish it from wick-based gaps, and evaluate it without making claims about invisible institutional positions. A complete EUR/USD example separates formation from a later entry, while a bearish example demonstrates failure. All prices, candles, and outcomes are hypothetical teaching constructions, not historical results or live recommendations.
Key insight: An ICT volume imbalance is a positive price separation between the bodies of two adjacent candles. The bodies do not touch or overlap, although the candle ranges can overlap through their wicks. Measure the nearest body edges using the open and close. The pattern is an ICT price-delivery reference, not a calculation of traded volume, buying pressure, or the number of unfilled orders.
1. What is an ICT volume imbalance?
A candle body spans its opening and closing prices. The wick extends to the highest and lowest recorded prices during the period. Those are different intervals. A volume imbalance exists under the adjacent-body definition when one candle’s entire body sits above or below the other candle’s body with a positive separation between them.
For an upward separation, the lower edge of the second body is above the upper edge of the first. For a downward separation, the upper edge of the second body is below the lower edge of the first. The candle colours help you locate the open and close, but they do not define the gap by themselves. Two green candles can have downward-separated bodies after a lower opening; two red candles can have upward-separated bodies after a higher opening.
ICT explains this focus on adjacent body separation in 2024 Mentorship Lecture 16, including the fact that wick activity does not automatically remove the body-based feature. This article applies that definition to completed candles so the measurements are fixed before an order is considered.
The distinction between completed and developing candles matters. A new candle may open above the previous body, then fall far enough that its eventual close makes the bodies overlap. If you classify the pattern from the opening tick alone, the final chart may no longer contain the setup you thought you traded. Waiting for completion is a rule chosen here to make the study reproducible.
Within ICT institutional order flow thinking, a volume imbalance is one of the price references used to interpret how a directional move was delivered. It can serve as a location to observe a retracement, an intermediate objective, or a supporting reference while a trade develops. Calling it institutional does not establish who traded there. The observable fact is the relationship between the two bodies; the order flow narrative is an interpretation added to that fact.
2. Why the word volume does not mean measured trade volume
You can calculate the ICT pattern from four prices: the open and close of each candle. No volume value is required. That alone tells you that the feature is different from a method comparing traded quantities at the bid and ask. A chart can contain a volume imbalance even when no volume histogram is displayed, and a volume spike can occur without any separation between adjacent bodies.
MetaTrader distinguishes price fields from volume fields. The official MqlRates reference lists open, high, low, close, tick volume, and trade volume separately. Those fields answer different questions. The ICT body-gap calculation uses the price fields. Adding tick volume to the chart does not convert that calculation into a measurement of institutional participation.
The platform’s Volumes documentation describes forex tick volume in terms of price changes during the selected period. It is not a total of all currency units traded worldwide. A large tick count may be useful in another research framework, but it is not part of the geometric definition taught here.
Nor does a body separation prove that no trading occurred between its boundaries. Wicks may already cover the interval. Different venues and price feeds may also record different activity. The safe description is narrower: the selected candles’ bodies leave a gap on this chart. You cannot infer the exact amount of resting liquidity, the identity of a buyer, or a requirement for price to revisit the level from that observation alone.
Keep your journal language specific. Instead of writing institutions left unfilled orders here, write adjacent bodies separated by 1.2 pips, wick ranges overlap, and the later response met or failed the chosen condition. You can still study the pattern through ICT’s delivery framework while keeping the evidence distinct from the explanation.
3. How to identify and measure the pattern
Start with body intervals
For each candle, define the body low as the smaller of its open and close, and the body high as the larger. This works regardless of candle colour. An upward volume imbalance exists when the second body low is strictly greater than the first body high. A downward volume imbalance exists when the second body high is strictly less than the first body low.
If the nearest body edges are equal, the bodies touch and the separation is zero. If the intervals overlap, there is no positive body gap under this rule. Do not rely on a thin white line between candles on the screen. Horizontal spacing, outlines, and chart scaling can create a visual impression that is absent from the actual prices. Read the numerical values in the platform’s data display.
Measure an upward separation
Take a hypothetical EUR/USD candle opening at 1.08416 and closing at 1.08473. Its body spans 1.08416 to 1.08473. The next candle opens at 1.08485 and closes at 1.08539, so its body spans 1.08485 to 1.08539. The volume imbalance lies between 1.08473 and 1.08485.
Width = (1.08485 − 1.08473) / 0.00010 = 1.2 pips. Midpoint = (1.08473 + 1.08485) / 2 = 1.08479. These are feature measurements, not yet a trade. A midpoint is a coordinate that a model may choose to observe or use for an order. It does not establish that the centre has a superior response rate.
Now inspect the full ranges. The first candle has a high of 1.08492 and a low of 1.08403. The second has a high of 1.08558 and a low of 1.08461. Their ranges overlap from 1.08461 to 1.08492. There is therefore no complete high-to-low gap between these adjacent candles, even though their bodies are separated.
Measure the downward version the same way
For a separate pair, let the first candle open at 1.09291 and close at 1.09237. Let the next open at 1.09225 and close at 1.09171. The nearest body edges are 1.09237 and 1.09225, producing a downward separation of (1.09237 − 1.09225) / 0.00010 = 1.2 pips. The midpoint is 1.09231. If the wick ranges overlap, it remains a body-gap feature rather than a complete adjacent-candle price gap.
Use the instrument’s permitted price increment when recording boundaries or an order price. A mathematical midpoint can fall between executable increments. If rounding is necessary, write down the rule before testing and recalculate the actual entry risk. Do not display an extra decimal place and assume the broker will accept it.
The body gap and the wick overlap are different intervals. In both examples, wick activity covers the VI even though the source bodies remain separate. A third candle is required for a separate FVG test; these two-candle measurements do not establish one.
4. Volume imbalance versus FVG, price gap, and volume analysis
A fair value gap uses three candles and compares the wick boundaries of the first and third. A bullish FVG exists when the third candle’s low is above the first candle’s high. A volume imbalance compares two adjacent bodies. The tests use different inputs, so one feature can exist without the other. They can also occur near each other during the same expansion.
In the upward measurement example, a later third candle with a low of 1.08469 would overlap the first candle’s high of 1.08492. That three-candle group would not form a bullish FVG under the stated wick test. The existing volume imbalance between the first two bodies would still be identifiable as a historical feature. Do not relabel it an FVG simply because a later price response looks attractive.
| Concept | What is measured? | What does it describe? | Key limitation |
|---|---|---|---|
| ICT volume imbalance | Nearest body edges of two adjacent candles. | A positive separation between open-close intervals. | Wicks can already cover the interval; no traded-volume quantity is measured. |
| ICT fair value gap | First and third candle wick boundaries in a three-candle sequence. | A specific non-overlap around the middle candle’s delivery. | The pattern does not guarantee a later return or successful entry. |
| Complete adjacent-candle price gap | The full high-low ranges of adjacent candles. | No overlap between those recorded ranges. | A gap on one feed is not proof of an absence of trading everywhere. |
| Tick volume | Price-update activity during a period. | A count supplied by the chart’s data source. | It is different from consolidated global traded quantity. |
| Bid-ask volume imbalance | Classified transaction quantities at prices or across a period. | An order flow measure requiring suitable transaction data. | It cannot be reconstructed reliably from two candle bodies alone. |
The official MetaTrader Depth of Market documentation distinguishes exchange transaction information from the broker-dependent information available for over-the-counter instruments. That is a useful reminder when moving between futures demonstrations and retail forex charts. Similar-looking screens can contain different kinds of data.
For ICT study, keep separate annotations when features coexist. Record the VI’s body boundaries and the FVG’s wick boundaries independently. Their midpoints need not match, and their later retests need not happen on the same candle. The dedicated fair value gap guide covers the three-candle construction in detail.
5. When a volume imbalance matters in an ICT trade
The feature becomes more useful when it answers a question already posed by the broader setup. If the context favours an attempt towards an old high, you might inspect a body separation created during bullish displacement for a possible retracement reference. If the context is unclear and price is overlapping in both directions, marking every small body gap does not supply the missing directional premise.
Locate the reference within the selected dealing range. Is it near a relevant liquidity event? Did it form during the displacement your model required? Does the route to the intended objective contain a nearby opposing array? Is the distance to that obstacle sufficient relative to structural invalidation? These questions connect the price feature to a decision rather than treating it as a standalone pattern to trade whenever it appears.
In 2024 Mentorship Lecture 34, ICT describes volume imbalances as comparatively flexible references that can allow price to move through them. The practical interpretation is to avoid treating every boundary as an impenetrable line. A later crossing may matter to a chosen entry model without erasing the original pattern from the historical chart.
This flexibility does not justify unlimited discretion. You still need a hard trade invalidation and a defined response to failure. A trader cannot argue that the imbalance should hold when entering, then claim that any penetration is acceptable after the position moves against them. Choose the rule first. If your plan uses a structural stop beyond a sweep extreme, calculate that full risk instead of pretending the narrow body-gap width is the stop distance.
Separate an entry reference from a target reference as well. A higher-timeframe VI can be an area towards which you expect price to travel, while a smaller-timeframe array supplies the entry. The same label can serve different roles, but the journal should state each role explicitly. The IRL and ERL guide explains how to place these references within named ranges.
6. Complete EUR/USD long example with risk and costs
Context and completed formation
Assume a hypothetical winter London morning. On the fifteen-minute context chart, the trader has a conditional bullish draw towards buy-side liquidity above an old high at 1.08771. The selected route has no additional material opposing array before the planned exit under this trader’s mapping rules. Price first trades below a local sell-side reference at 1.08414 to an extreme of 1.08401.
All times below identify five-minute candle closes in Ghana GMT. The candle closing at 8:05 AM makes the sweep and closes at 1.08416. The 8:10 AM candle opens at 1.08416 and closes at 1.08473, with high 1.08492 and low 1.08403. The 8:15 AM candle opens at 1.08485 and closes at 1.08539, with high 1.08558 and low 1.08461.
These are the two source candles from the measurement section. Their completed bodies define the volume imbalance from 1.08473 to 1.08485. The second close at 1.08539 also exceeds the preselected internal high of 1.08517. Under the rules used here, the sweep, bullish displacement close, and completed body separation jointly authorise evaluation of a retracement entry.
Notice what is already true at formation. The second source candle’s lower wick has passed below the entire body-gap interval. This does not invalidate the body-separation measurement. It does mean that the zone must not be described as an untouched range with no previous price activity. It also cannot be counted as a successful later entry, because the completed pattern was not available at the start of that candle.
The order is placed only after confirmation
After the 8:15 AM candle closes, the trader places a buy limit at the midpoint, 1.08479. The protective stop is 1.08387, which is 1.4 pips below the sweep extreme of 1.08401. The full-position target is 1.08755, which is 1.6 pips before the old high at 1.08771. The order expires at 8:45 AM if unfilled and is cancelled sooner if the target trades first or the context becomes invalid before entry.
The candle closing at 8:20 AM reaches a low of 1.08469 and closes at 1.08504. For this constructed example, assume the broker’s executable ask reaches the buy-limit price and the order fills at 1.08479. The wick passes slightly below the VI’s lower boundary, while the completed body remains above it. The plan’s invalidation is the stated structural stop, not an automatic exit on that wick penetration.
Price then expands upward, pauses once, and reaches 1.08763 during the candle closing at 8:50 AM. The planned target at 1.08755 is reached first in this illustrative path. This describes the desired outcome of the selected rules. It does not estimate their win rate or demonstrate that a small VI caused the later move.
The source candle has already wicked through the body-gap interval while forming. That earlier activity does not count as a later entry. The diagram projects the completed VI for reference and assumes the broker ask reaches the buy limit on the later retest. The separate bearish example demonstrates a stop-first loss.
Calculate the actual trade distance
Long trade arithmetic: Risk = (1.08479 − 1.08387) / 0.00010 = 9.2 pips. Reward = (1.08755 − 1.08479) / 0.00010 = 27.6 pips. Gross reward divided by risk = 27.6 / 9.2 = 3.00R, giving a risk-to-reward ratio of 1:3. The VI is 1.2 pips wide, but the trade risks 9.2 pips before costs.
Use a simplified all-in round-trip cost allowance of 0.8 pip for planning. Net target reward is approximately 27.6 − 0.8 = 26.8 pips. Planned stopped loss is approximately 9.2 + 0.8 = 10.0 pips. The cost-adjusted reward multiple is 26.8 / 10.0 = 2.68R. Apply the allowance once per completed outcome; it is not a charge to add independently at every intermediate price reference.
For a hypothetical USD 2,000 account risking 0.5%, the cash budget is USD 10. Assuming EUR/USD uses a 100,000-euro standard lot and an approximate USD 10 pip value in a USD account, position size = 10 / (10.0 × 10) = 0.10 standard lot. The estimated stop outcome is 10.0 × USD 1 = USD 10 lost, and the target outcome is 26.8 × USD 1 = USD 26.80 gained under those assumptions.
Actual fills depend on the executable quote and broker handling. A bid-only chart touching a buy limit does not establish that the ask reached it. The MetaTrader trading principles explain the distinction between buy and sell execution prices. If your actual fill calculation already includes spread, do not add the same spread again. Slippage and variable costs can make the stopped loss larger than planned.
For a Ghana-based trader using a cedi risk budget, convert the budget into the account currency before calculating the order size. Use the broker’s contract specifications and permitted lot increment. The small visual width of the VI does not reduce the monetary risk unless a tested model independently supports a closer valid stop.
7. A bearish example that fails after entry
Now use the separate downward body separation from 1.09225 to 1.09237. Before those candles form, a hypothetical EUR/USD move trades above a local high at 1.09299 to a sweep extreme of 1.09315. The first source candle opens at 1.09291 and closes at 1.09237. The next opens at 1.09225 and closes at 1.09171, below the preselected internal low of 1.09203.
The midpoint is (1.09225 + 1.09237) / 2 = 1.09231. After completion, the trader places a sell limit there, a protective stop at 1.09329, and a target at 1.08937 before a selected sell-side reference at 1.08921. The stop is 1.4 pips above the sweep extreme. The context, displacement, and target map are separate qualifications; the downward body gap alone would not authorise the short.
Short calculation: Risk = (1.09329 − 1.09231) / 0.00010 = 9.8 pips. Reward = (1.09231 − 1.08937) / 0.00010 = 29.4 pips. Gross reward-to-risk = 29.4 / 9.8 = 3.00R, or 1:3 risk-to-reward. With the same 0.8-pip round-trip allowance, potential net reward is 28.6 pips and planned stopped loss is 10.6 pips. The adjusted multiple is 28.6 / 10.6 = 2.70R.
In the failed path, a later retracement fills the sell limit, but renewed buying carries price to the stop at 1.09329 before the target trades. The realised result is a 9.8-pip loss before costs, or approximately 10.6 pips including the stated allowance. The source candles still had a valid volume imbalance. What failed was the trade hypothesis built around that feature.
The trader does not widen the stop because a VI can be revisited later. Pattern persistence and trade validity are different ideas. A historical annotation may remain useful for study after a stop-out, but a new trade requires new evidence and a new risk calculation. Reusing the old narrative to keep increasing exposure would defeat the purpose of structural invalidation.
Also distinguish this loss from a cancelled idea. If the pattern never completes, if the required structure confirmation is absent, or if the target is reached before the pending order fills, the rules produce no trade. A study should record all of those categories. Counting only attractive target-first examples would exaggerate what the model actually delivered.
8. Feed precision, tiny zones, and chart construction
Small body gaps are sensitive to the prices used to build the chart. One provider’s final quote for a candle and first quote for the next may differ from another provider’s observations. Session boundaries, missing records, quote conventions, and rounding can also affect apparent separation. These are data questions to investigate, not automatic evidence that the pattern is meaningless or that the broker is manipulating your chart.
Use standard price candles for this measurement. Synthetic candle constructions change the relationship between displayed opens and closes and the underlying traded or quoted prices. A body gap calculated from transformed candles is not the same test as one calculated from ordinary OHLC bars. Keep the chart type fixed during both discovery and evaluation.
If the next open equals the previous close exactly, those two body intervals share that price and cannot have a strictly positive separation. This follows directly from the definition. When you see many supposed VIs on such a chart, inspect the numerical boundaries. You may be identifying visual spacing, a different type of inefficiency, or a rule that compares non-adjacent candles instead.
For the EUR/USD measurement example, the gap is 0.00012 wide. On a five-decimal quote display, that is 12 displayed points of 0.00001, equal to 1.2 pips. A pip and the smallest displayed point are not interchangeable. Confusing them can make an entry buffer or a position-size calculation ten times larger or smaller than intended.
Confirm the actual tick size and contract settings in the symbol specification rather than inferring them from the number of digits. The official Market Watch documentation describes chart mode, tick size, tick value, and related symbol information. These details matter when comparing a body gap with the prices at which orders can execute.
A zone can also be too small to offer meaningful execution precision. Compare its width with normal spread variation and the uncertainty in your fill assumptions. That comparison does not prove a trade will lose, because its stop and target may be much farther away. It tells you whether choosing the near edge, midpoint, or far edge is a distinction your execution data can actually support.
9. Ghana timing and a repeatable session workflow
The clock determines when the selected model is active, not whether two bodies are separated. A VI can appear outside your trading hours and remain a reference on the chart. That does not require you to place an order. Use the session plan to limit attention, then require the same price and risk conditions you would apply at any other permitted time.
Ghana remains on GMT year-round. New York shifts between standard time and daylight saving time. The familiar London Kill Zone of 7 to 10 AM Ghana time and New York Kill Zone of noon to 3 PM Ghana time are the winter conversions of the New York-anchored windows used here. The Ghana equivalents move one hour earlier during New York daylight saving time.
| Reference in New York local time | Ghana during New York standard time | Ghana during New York daylight saving time |
|---|---|---|
| London Kill Zone, 2 to 5 AM | 7 to 10 AM GMT | 6 to 9 AM GMT |
| New York Kill Zone, 7 to 10 AM | Noon to 3 PM GMT | 11 AM to 2 PM GMT |
| New York Midnight Open, 12:00 AM | 5 AM GMT | 4 AM GMT |
Verify dates using the Ghana time-zone reference and New York clock reference. London local opening time is a separate convention because the UK and US clock changes can occur in different weeks. Broker server time may differ again. Record whether your chart timestamps show candle opens or closes.
Before the session, mark the context range, likely objectives, and scheduled releases. During the session, observe whether the chosen liquidity event and displacement occur. After a candidate VI completes, calculate its boundaries and assess a later entry. If the order is not filled before its deadline, cancel it. This sequence prevents a body gap from becoming a permanent invitation to trade whenever price eventually returns.
Do not treat news-driven separation as automatically favourable. A quote jump can make a feature more visible while also making execution less predictable. If the model excludes new exposure around major releases, respect that rule even when the chart looks unusually clean. The relevant question is whether the conditions match the tested plan.
10. How to test volume imbalances without hindsight
Begin with a fixed detector definition: adjacent completed bodies, strict positive separation, and standard candles from one named feed. Record upward and downward separation separately from candle colour. Save the open, close, high, low, timeframe, and completion time of both source candles. These records let you verify that the pattern existed before the entry was possible.
Record wick overlap independently. A VI with overlapping full ranges is different from a body separation inside a complete price gap. Also note whether a separately measured FVG is present. Do not combine all three categories under a single imbalance label, because a result driven by one category could otherwise be mistakenly attributed to another.
Next, record the role assigned to the VI: entry reference, target, or observation only. Define the surrounding conditions and the exact order rules. The long example used a prior sweep, a completed displacement close, a midpoint limit, a structural stop, and a fixed target. An edge-entry model with a candle-close exit is a different strategy and should have a separate test.
For outcomes, keep unfilled orders, cancelled orders, stop-first trades, target-first trades, and ambiguous sequences. If one historical bar spans both entry and target, do not assume the target traded after the fill. If a bar spans both stop and target, use sufficient lower-level data or a predetermined conservative convention. Candle extremes alone do not reveal the order of every price visit.
The MetaTrader guide to real and generated ticks explains that historical testing can use different data and generation methods. For a feature only a few points wide, those assumptions deserve explicit attention. A backtest with idealised fills may describe a price pattern accurately while overstating what an executable strategy could earn.
Finally, test whether the VI contributes information beyond the existing setup. Compare the same entry framework with and without the additional VI condition, keeping other rules stable. Include costs and enough independent observations to avoid drawing conclusions from a few examples. A useful result may be that the feature improves location selection, helps define a target, or adds no measurable benefit under your conditions. All three answers are more informative than a collection of attractive screenshots.
11. Frequently asked questions
What is a volume imbalance in ICT trading?
It is a positive separation between the bodies of two adjacent candles. The nearest open-close edges do not touch or overlap. The wick ranges may overlap. ICT uses the feature as a price-delivery reference, but the measurement itself does not calculate actual buying or selling volume.
Can the wicks overlap and still form a VI?
Yes. The defining comparison is between the bodies. In the upward example, the bodies leave a gap from 1.08473 to 1.08485 even though the full ranges overlap. This is why the feature must not be described as an interval where price has never traded. Wick activity and body separation answer different questions.
Does candle colour determine whether the imbalance is bullish or bearish?
No. The direction of separation depends on the relative body intervals, not colour alone. Two green candles can be separated downward after a lower opening. Distinguish the geometric direction of the gap from the bullish or bearish trade thesis, which requires context and confirmation.
Is a volume imbalance the same as a fair value gap?
No. A VI compares two adjacent candle bodies. A FVG uses three candles and compares the first and third candle wick boundaries. The features can coexist, but one does not prove the other. Mark their boundaries separately and use the rules assigned to the actual feature you are trading.
Do I need a volume indicator to find it?
No. The calculation requires the open and close of two candles. Tick volume, traded volume, and bid-ask volume analysis are separate data concepts. You may study them in another model, but they are not required to identify adjacent body separation under the definition used here.
Should I always enter at the midpoint?
No. The midpoint is an exact coordinate, not a proven best entry. This article uses it to make the examples testable. A different plan might use an edge or wait for an additional response. Compare the alternatives with realistic fills, and round only according to the instrument’s permitted price increment.
Does a wick through the VI invalidate a trade?
That depends on the predetermined trade rules. A wick through the interval does not erase the original body separation. A particular strategy may still treat that penetration as invalidation. The worked long instead uses a structural stop below the sweep. Do not switch between these rules after entering to avoid accepting a loss.
Why is the feature different on two brokers’ charts?
Feeds, quote conventions, session boundaries, and recorded opens or closes can differ. Small body gaps are sensitive to these differences. Check the numerical OHLC values and symbol settings before drawing a conclusion. Test the pattern on the feed and execution conditions relevant to the account you intend to use.
Can a valid volume imbalance still produce a losing trade?
Yes. A valid geometric feature does not guarantee that the directional hypothesis will succeed. The bearish example has a defined body gap and confirmation but reaches its protective stop first. Separate pattern detection, order execution, and trade outcome in your records so a losing result does not lead to retrospective rule changes.
Place the VI within a complete ICT framework by studying the fair value gap construction and the role of an order block in the wider price sequence. Use the London Kill Zone walkthrough to organise timing, the market structure shift guide to define confirmation, and the Judas swing explanation to examine the initial liquidity event. Measure the actual body edges, then make the entry, invalidation, and target precise enough to evaluate.